Method and system for automatically calibrating a color display

FIELD OF THE INVENTION

0001The present invention relates in general to displaying images on a color display device and in particular to a method and system for automatically adjusting and correcting luminance and color chromaticity values to improve the perceived image quality of the color display device.

BACKGROUND OF THE INVENTION

0002Color images can be captured and converted into a video signal, which can be transmitted to a display system, such as a television. The display system typically processes the input video signal and transmits the processed video signal to a display device, which reproduces the luminance and color of the images onto its display screen for a viewer. Typical display devices include liquid-crystal displays (LCD), cathode-ray tubes (CRT), and plasma display panels (PDP). Each display device employs complex mechanisms that take the video signal and reproduce the luminance and color properties of the image.

0003For example, the luminance and color reproduction of an LCD is characterized by the spectral power distribution (SPD) of a backlight, e.g., cold-cathode fluorescent light (CCFL) tubes or light-emitting diodes (LEDs), the transmission characteristics of the polarizers and color filters for each primary-color sub-pixel on the screen, and the transmission characteristics of the liquid crystal cells under different electrical field strengths controlled by each pixel data from the input video signal. The luminance and color reproduction of a CRT are characterized by the SPD of each kind of phosphor material of each primary-color dot in a triad on the screen and the electron emitting dynamics of the electron guns under different electrical field strengths on the grids controlled by the input video signal. The luminance and color reproduction of a PDP are characterized by the SPD of each kind of phosphor material of each primary-color sub-pixel on the screen and the electrical discharge dynamics of the dielectric layers among the electrodes with voltage differences controlled by each pixel data from the input video signal.

0004Important properties of a display device are characterized by the chromaticity values of the device's primary color components, the reference white point, and the device's power transfer function from input signal voltage to output luminance level. In a typical display device, luminance curves, gamma values, color chromaticity values, and color temperatures are set to predetermined nominal settings during the manufacturing process. With these settings, the ideal display device can accurately reproduce the luminance and color properties of the captured image to provide an enjoyable viewing experience for the viewer.

0005Nonetheless, most mass-produced display devices are less than ideal. The predetermined nominal settings are difficult to attain due to the display device's imperfections that arise during the manufacturing process, as well as the complex underlying physical mechanisms of the display devices themselves. Thus, for standard mass-produced display devices, where quality control during manufacturing is more lax in order to keep costs low, the predetermined nominal settings are not necessarily achievable. For instance, because the nominal settings are usually mismatched, it is common for standard display devices to exhibit undesirable gray level color deviations and color temperature shifts. Such gray-level color deviations cause the measured chromaticity values of a displayed white color to drift on a chromaticity diagram with different gray levels. Moreover, gray-level color deviations also cause the measured color temperature of a displayed white color to vary with different gray levels. These undesirable color deviations and color temperature shifts diminish the image quality of the display device.

0006Moreover, the actual, i.e., measured, luminance curve of the standard display device can also deviate from the predetermined power-law transfer function. Accordingly, the measured luminance curves, gamma values, color chromaticity values, and color temperatures of typical mass-produced display devices usually deviate from expected values, and in some instances, the values that should be constant, e.g., gamma values, are variable and curves that are supposed to be smooth, e.g., luminance curves, are uneven. For these reasons, standard mass-produced display devices often render non-ideal perceived image quality.

0007Accordingly, it is desirable to provide a method and system for improving the perceived image quality of color display devices. In particular, it is desirable to provide a method for automatically calibrating a color display device such that the output characteristics of the display device are substantially in line with the expected output characteristics of an ideal display device.

SUMMARY OF THE INVENTION

0008In one version, a system for calibrating a display device to improve its perceived image quality includes a calibration module that is configured to determine, for each of a plurality of white colors associated with a plurality of gray levels displayed sequentially by the display device, a measured chromaticity point on a chromaticity diagram and a measured luminance level. The calibration module calculates, for each gray level, a differential change in each primary color component that simultaneously moves the measured chromaticity point to a target chromaticity point and adjusts the measured luminance level to a target luminance level on a predetermined luminance curve having a target gamma value, and calculates correction values for each primary color component and each gray level based on the calculated differential changes. The system also includes means for outputting the calculated correction values to the display device. The display device corrects the primary color components of a color video signal based on the calculated correction values such that the display device accurately reproduces luminance and color properties of the color video signal.

0009In another version, a display device includes a display screen that displays primary color components of a color video signal and a plurality of lookup tables. Each lookup table is associated with a primary color component and loads correction values for the associated primary color component. Each table includes an output that is coupled to an input of the display screen. Each of the plurality of lookup tables receives the associated primary color component of an input color video signal and outputs the correction value to the display screen so that the display device accurately reproduces luminance and color properties of the color video signal.

0010In another version, a display system includes a display device that displays primary color components of an input color video signal and the plurality of lookup tables that are loaded with the correction values for the associated primary color components.

0011In another version, a method for improving the perceived image quality of a display device used for displaying a color video signal includes determining, for each of a plurality of white colors associated with a plurality of gray levels displayed sequentially by the display device, a measured chromaticity point on a chromaticity diagram and a measured luminance level, calculating, for each gray level, a differential change in each primary color component that simultaneously moves the measured chromaticity point to a target chromaticity point and adjusts the measured luminance level to a target luminance level on a predetermined luminance curve having a target gamma value, and adjusting the primary color components of the color video signal based on the calculated differential changes such that the display device accurately reproduces luminance and color properties of the color video signal.

DESCRIPTION OF THE DRAWINGS

0012These features, aspects and advantages of the present invention will become better understood with regard to the following description, appended claims, and accompanying drawings, which illustrate examples of the invention. However, it is to be understood that each of the features can be used in the invention in general, not merely in the context of the particular drawings, and the invention includes any combination of these features, where:

0013<figref idref="DRAWINGS">FIG. 1</figref> is a CIE chromaticity diagram;

0014<figref idref="DRAWINGS">FIG. 2</figref> is a typical luminance curve for a display device;

0015<figref idref="DRAWINGS">FIG. 3</figref> is a block diagram of the calibration system coupled to a display device according to a version of the present invention;

0016<figref idref="DRAWINGS">FIG. 4</figref> is a graph illustrating the chromaticity trajectory of a display device that has not been calibrated;

0017<figref idref="DRAWINGS">FIG. 5</figref> is a flowchart illustrating a calibration process performed by the calibration system according to a version of the present invention;

0018<figref idref="DRAWINGS">FIG. 6A</figref> and <figref idref="DRAWINGS">FIG. 6B</figref> are graphs showing initial values for the three intermediate gamma correction LUTs corresponding to red, green, and blue primary color components versus gray level and the measured luminance levels before calibration and a target luminance curve versus gray level, respectively;

0019<figref idref="DRAWINGS">FIG. 7A</figref> and <figref idref="DRAWINGS">FIG. 7B</figref> are graphs showing updated values for the three intermediate gamma correction LUTs, i.e., LUT_R(i), LUT_G(i), and LUT_B(i), versus gray level and updated luminance levels Y_s(i) versus gray level;

0020<figref idref="DRAWINGS">FIG. 8A</figref> and <figref idref="DRAWINGS">FIG. 8B</figref> are graphs showing adjusted values of the three intermediate gamma correction LUTs versus gray level and the luminance levels after calibration versus gray level, respectively;

0021<figref idref="DRAWINGS">FIG. 9</figref> is a graph showing the chromaticity trajectory after the calibration process according to a version of the present invention;

0022<figref idref="DRAWINGS">FIG. 10</figref> is a flowchart illustrating a calibration process according to another version of the present invention;

0023<figref idref="DRAWINGS">FIG. 11</figref> illustrates a three-dimensional luminance-chromaticity diagram;

0024<figref idref="DRAWINGS">FIG. 12</figref> depicts an exemplary display system according to a version of the present invention;

0025<figref idref="DRAWINGS">FIG. 13</figref> depicts a display screen containing multiple display regions with different input video signal characteristics; and

0026<figref idref="DRAWINGS">FIG. 14</figref> depicts an exemplary display system according to another version of the present invention.

DESCRIPTION OF THE INVENTION

0027The present invention relates in general to color display devices that display a color video signal and in particular to a method and system for automatically adjusting and correcting luminance and color chromaticity values to improve the perceived image quality of the color display. The following description is presented to enable one of ordinary skill in the art to make and use the invention and is provided in the context of a patent application and its requirements. Various modifications to the preferred embodiments and the generic principles and features described herein will be readily apparent to those skilled in the art. For instance, although the calibration system and process according one version of the present invention uses the CIE (X, Y, Z) tristimulus values and (x, y) chromaticity values, the method and system of the present invention is not necessarily limited to the CIE XYZ color system and the derived (x, y) chromaticity values. Those skilled in the art would readily appreciate that other well-defined color systems can also be used according to the method and system of the present invention, notably the perceptually uniform CIE (u′, v′) chromaticity values or the perceptually uniform CIE L*u*v* or CIE L*a*b* color systems. Thus, the present invention is not intended to be limited to the embodiments shown, but is to be accorded the widest scope consistent with the principles and features described herein.

0028The characteristics of color inherently distinguishable by the human eye are chromaticity values (related to hue and saturation), and luminance (related to brightness). Accordingly, color systems characterize colors in various parameters which relate to hue, saturation, and brightness. Such a system includes the quantitative XYZ color system defined by Commission Internationale de l'Éclairage (CIE) in 1931, where all three tristimulus values (X, Y, Z) are positive and all visible colors are unambiguously represented by two chromaticity values (x, y) derived from the three tristimulus values (X, Y, Z). A mapping of all the visible colors produce a shark-fin-shaped region on the (x, y) plane known as a CIE (x, y) chromaticity diagram (illustrated in <figref idref="DRAWINGS">FIG. 1</figref>). The shark-fin-shaped region on the chromaticity diagram represents the whole range of human color perception. All the monochromatic colors are distributed around the curved edge of the region, defined by a spectral locus that is joined by a line of purples. All perceivable colors lie within the shark-fin-shaped region on the chromaticity diagram. Points outside the region are not representing colors to the human eye. The triangle located within the region and defined by the three primary colors, red, green and blue (R, G, B), represents the perceived colors that can be matched by mixing the three primary colors. The extent, or gamut, of the colors that can be matched by mixtures from a given set of three primary colors is given on the chromaticity diagram by a triangle whose vertices are the chromaticity values of the three primary colors.

0029Any color on the CIE (x, y) chromaticity diagram can be considered to be a projection from the three-dimensional space of the three CIE tristimulus values (X, Y, Z) to the two-dimensional plane of the two CIE chromaticity values (x, y). The CIE tristimulus values (X, Y, Z), which are always non-negative, can represent all perceivable colors. In addition, the Y tristimulus value determines the luminance of a light source, while the (x, y) chromaticity values determine the color of a light source. The CIE tristimulus values (X, Y, Z) can be derived from the spectral power density (SPD) of a colored object using the three corresponding color matching functions defined in the wavelength domain. Different combinations of light wavelengths that result in the same set of CIE tristimulus values will be indistinguishable to the human eye.

0030Another characteristic of color inherently distinguishable by the human eye is known as the color temperature of a light source. Color temperature is characterized in color science by the temperature, in degrees Kelvin (K), of a black-body radiator which radiates the color light with the same color perceived by the human eye as the white light from a given source. Correlated color temperature (CCT) extends the idea to include light sources that can be closely but not exactly matched by a black-body radiator. Referring again to <figref idref="DRAWINGS">FIG. 1</figref>, a Planckian (white) locus represents temperatures of white ranging from about 1500K to about 10000K. For example, early morning daylight has a correlated color temperature of about 3000K (D30) and has a relatively reddish tone. An overcast midday sky has a correlated color temperature of about 10000K (D100) and has a relatively bluish tone. In the discussions that follow, correlated color temperature is implied when the term color temperature is used.

0031Through observation, it is known that the brightness and color temperature of the ambient light of the viewing environment and the brightness and color temperature of a displayed image affects the perceived picture quality of a display device. To provide the optimum perceived picture quality, the color temperature of the displayed image on the screen can be adjusted in accordance with the brightness and color temperature of the ambient light. One known method of improving perceived picture quality is to equalize a displayed color image based on the color temperature of the ambient light by adjusting the intensities of the individual primary color components used in the display device (e.g., RGB) based on the color temperature of the ambient light.

0032In addition, subjective test results indicate that the human eye can distinguish differences in brightness, i.e., luminance, down to about 1% to 2% of the diffuse white, which refers to the luminance of a diffusely reflecting white surface in a scene, and that the human eye is most acutely sensitive to changes in brightness in dark images. Tests reveal that contrast sensitivity and luminance discrimination thresholds are best modeled by a nonlinear function such as a logarithmic or a power-law transfer function having a gamma value (γ) of less than one.

0033In world-wide video standards for analog television, e.g., ITU-R BT.470, and digital television, e.g., ITU-R BT.709, a power-law transfer function having a gamma value of between 0.4 and 0.5 is assumed in the video coding at the capture source, i.e., the camera. After applying the nonlinear power-law transfer function, the analog or digital electronic video signal is referred to as gamma pre-corrected signals. The gamma value, γ, of the display device for which the gamma pre-corrected video signal is intended to be displayed is assumed to be approximately between 2.2 and 2.8. The gamma value of the display device is defined as the slope of the logarithm of the luminance curve as a function of the logarithm of the input video signal amplitude when the brightness control of the display is set so as to make the luminance curve as straight as possible over a luminance range corresponding to a contrast as high as possible. <figref idref="DRAWINGS">FIG. 2</figref> illustrates a typical luminance curve for a display device, where the curve, f(x), is characterized by: <br /><i>f</i>(<i>x</i>)=<i>x</i>^γ

0034As stated above, most typical display devices are preconfigured by the device manufacturer to have luminance curves, gamma values, color chromaticity values, and color temperatures set to predetermined nominal settings. These predetermined nominal settings are not necessarily device specific and are difficult to attain due to variations in manufacturing practices, such as quality control, that can affect the underlying display mechanisms of the device. Thus, the measured luminance curves, gamma values, color chromaticity values, and color temperatures of typical mass-produced display devices often do not reflect the expected values. Rather, they typically are inaccurate and inconsistent; and the luminance curves exhibit uneven behavior, which is highly undesirable.

0035To address this serious concern, a version of the present invention provides a calibration system that calculates correction values that can be used to adjust the input video signal such that a measured luminance curve of a display device has a desired predetermined gamma value and a displayed color has predetermined chromaticity values and color temperature. In one version, the calibration system can calculate more than one set of correction values so that the display device can be characterized by more than one gamma value and more than one set of chromaticity values and color temperature. This is desirable in order to achieve optimum perceived image quality under different viewing conditions and input video signal characteristics.

0036In a preferred embodiment, the calibration system calculates, for a plurality of gray levels above a threshold gray level, how much the relative intensity of each primary color component used by the display device should be adjusted in order to fix chromaticity values to a predetermined target point on the chromaticity diagram and to fix the gamma value of the luminance curve to a target gamma value simultaneously. The adjustment values are used to calculate correction values that are loaded into a plurality of lookup tables in the display device and are used to correct an input video signal during operation.

0037<figref idref="DRAWINGS">FIG. 3</figref> is a block diagram of the calibration system <b>100</b> coupled to a display device <b>10</b> according to a version of the present invention. According to one version of the present invention, the display device <b>10</b> includes a plurality of embedded gamma correction LUTs <b>12</b> having outputs coupled to a display screen <b>14</b> and a storage mechanism <b>16</b> coupled to the embedded gamma correction LUTs <b>12</b>. In another version, the plurality of embedded gamma correction LUTs <b>12</b> can be external to the display device <b>10</b>, e.g., in a video processor module of a display system (not shown) that utilizes the display device <b>10</b>, such as a television. Each primary color component is associated with one of the plurality of embedded LUTs <b>12</b>. Correction values are loaded from the storage mechanism <b>16</b> into the embedded LUTs <b>12</b> and are used to adjust an input video signal so that the display device accurately reproduces the luminance and color of the input image. The correction values are determined by the calibration system <b>100</b> during a calibration process.

0038The calibration system <b>100</b> includes a measuring probe <b>110</b>, a calibration module <b>120</b> and a test pattern controller <b>130</b>. The test pattern controller <b>130</b> is capable of generating a video signal corresponding to a known test pattern, and is directly coupled to the display device <b>10</b>, such that the display device <b>10</b> displays the input video signal, i.e., the test pattern, on the display screen <b>14</b>. The test pattern controller <b>130</b> is capable of generating specific primary color (R, G, B) component values. In one version, the test pattern controller <b>130</b> can be a computer graphics card known in the art. Alternatively, the test pattern controller <b>130</b> can be a known pattern generator. The video signal generated by the test pattern controller <b>130</b> can be received directly by the display screen <b>14</b>, thereby bypassing the embedded LUTs <b>12</b>. In another version, the video signal can be inputted into the embedded LUTs <b>12</b>.

0039The measuring probe <b>110</b> is capable of measuring luminance and color property data of the light emitted from the surface of the display screen <b>14</b> when it displays the input video signal, i.e., the test pattern. The measuring probe <b>110</b> is coupled to the calibration module <b>120</b>, which includes a color analyzer <b>122</b> and a lookup table (LUT) value generator <b>124</b>. The color analyzer <b>122</b> of the calibration module <b>120</b> analyzes the luminance and color property data and calculates luminance levels and color tristimulus values corresponding to the measured data. The LUT value generator <b>124</b> includes a plurality of intermediate gamma correction LUTs <b>126</b> that store intermediate LUT values. In a preferred embodiment, each primary color component is associated with one of the plurality of intermediate gamma correction LUTs <b>126</b>. During the calibration process, the LUT value generator <b>124</b> uses the measured luminance levels and color tristimulus values to calculate appropriate correction values for the embedded gamma correction LUTs <b>12</b>.

0040In a calibrated system, the measured chromaticity point on the chromaticity diagram corresponding to all gray levels above a threshold gray level should fall on or near a predetermined target chromaticity point on the white locus. Below the threshold gray level, which is approximately 10% to 20% of the maximum gray level, the behavior of the measured chromaticity points is more erratic and less predictable. The behavior of the display screen <b>14</b> below the threshold gray level is less controllable by its inputs from the embedded gamma correction LUTs <b>12</b>, therefore, the calibration process is not effective in these gray levels. In an uncalibrated system, the measured chromaticity points on the chromaticity diagram corresponding to the gray levels above the threshold gray level are scattered, as illustrated in <figref idref="DRAWINGS">FIG. 4</figref>, and the measured color temperature can deviate from the target color temperature by as much as 5,000K. This is referred to as gray-level color deviation, which significantly degrades the perceived image quality of the display device.

0041According to a version of the present invention, the values for the embedded gamma correction LUTs <b>12</b> are calculated by analyzing the measured luminance levels Y and chromaticity values (x, y) corresponding to a plurality of gray levels and determining for each gray level a corresponding measured chromaticity point on the chromaticity diagram. The LUT value generator <b>124</b> determines, for each measured chromaticity point, the differential change in each primary color component value needed to move the measured chromaticity point to the target chromaticity point. The LUT value generator <b>124</b> then calculates the correction value of each primary color component for each gray level so that the target chromaticity values and the target luminance levels are matched throughout the gray levels above the threshold simultaneously. When this is completed, the display device <b>10</b> accurately reproduces the luminance and color properties of the input signal and the perceived image quality is optimized.

0042<figref idref="DRAWINGS">FIG. 5</figref> is a flowchart illustrating the calibration process performed by the calibration system <b>100</b> shown in <figref idref="DRAWINGS">FIG. 3</figref> according to a version of the present invention. Referring to <figref idref="DRAWINGS">FIG. 3</figref> and <figref idref="DRAWINGS">FIG. 5</figref>, the calibration process begins by initializing the intermediate gamma correction LUTs <b>126</b> with a linearly increasing function: <br />LUT_R(<i>i</i>)=<i>i, </i><br />LUT_G(<i>i</i>)=<i>i, </i><br />LUT_B(<i>i</i>)=<i>i, </i><br /> where i=0, 1, . . . , M−1 and M is the number of gray levels for each of the R, G, and B sub-pixels of the display screen <b>14</b> (step <b>500</b>). Typically, for 8-bit pixel data, M is equal to 256, and for 10-bit data, M is equal to 1024. <figref idref="DRAWINGS">FIG. 6A</figref> is a graph showing the initial values for the three corresponding intermediate gamma correction LUTs versus gray level, i.

0043After the intermediate gamma correction LUTs <b>126</b> have been initialized, the test pattern controller <b>130</b> generates R, G, and B component values corresponding to a plurality of test patterns and transmits the test patterns to the display device <b>10</b>, which displays the test patterns on the screen <b>14</b> (step <b>502</b>). Each displayed test pattern is preferably a white color corresponding to a selected gray level, k.

0044According to one version, the output of the test pattern controller <b>130</b> can be connected to the inputs of the embedded gamma correction tables <b>12</b>. In this configuration, the values of the intermediate gamma correction LUTs <b>126</b>, e.g., LUT_R(i), LUT_G(i), LUT_B(i), can be loaded into the corresponding embedded gamma correction LUTs <b>12</b> and the test pattern controller <b>130</b> can be directed to generate R, G, and B component outputs (k, k, k), where k are selected gray levels. Alternatively, the values of the embedded gamma correction LUTs <b>12</b> can be (i, i, i), where i=0, 1, . . . , M−1, and the test pattern controller <b>130</b> can be directed to generate R, G, and B component outputs corresponding to the values of the intermediate gamma correction LUTs <b>126</b> for k, e.g., LUT_R(k), LUT_G(k), LUT_B(k), where k are selected gray levels.

0045According to another version, the output of the test pattern controller <b>130</b> can be connected directly to the inputs of the display screen <b>14</b>. In this configuration, the test pattern controller <b>130</b> can be directed to generate R, G, and B component outputs corresponding to the values of the intermediate gamma correction LUTs <b>126</b> for k, e.g., LUT_R(k), LUT_G(k), LUT_B(k), where k are selected gray levels.

0046For every displayed test pattern, e.g., a displayed white color corresponding to the selected gray level, the luminance level Y(k) and chromaticity values [x(k), y(k)] are determined (step <b>504</b>). In particular, the measuring probe <b>110</b> measures the luminance and color property data of the displayed test pattern and the color analyzer <b>122</b> converts the luminance and color property data into the CIE tristimulus values (X, Y, Z), which in turn are used to calculate the chromaticity values (x, y). <figref idref="DRAWINGS">FIG. 6B</figref> is a graph showing the measured luminance curve before calibration and a target luminance curve versus gray level. In one version, the calibration system <b>100</b> determines the following values for each selected gray level k: <br /><i>Y</i>(<i>k</i>)=<i>Y[LUT</i>_R(<i>k</i>), <i>LUT</i>_G(<i>k</i>), <i>LUT</i>_B(<i>k</i>)], 0≦<i>k≦M−</i>1;<br /><i>x</i>(<i>k</i>)=<i>x[LUT</i>_R(<i>k</i>), <i>LUT</i>_G(<i>k</i>), <i>LUT</i>_B(<i>k</i>)], 0≦<i>k≦M−</i>1;<br /><i>y</i>(<i>k</i>)=<i>y[LUT</i>_R(<i>k</i>), <i>LUT</i>_G(<i>k</i>), <i>LUT</i>_B(<i>k</i>)], 0≦<i>k≦M−</i>1;<br /> where Y[•], x[•], and y[•] are the unknown underlying functions of the three component values at the inputs of the display screen <b>14</b> in the display device <b>10</b> under test.

0047The LUT value generator <b>124</b> receives the measured luminance levels and chromaticity values and compares each set to a predetermined target luminance level Y_T(k) and a predetermined target chromaticity point (x_T, y_T) (step <b>506</b>) to determine whether the deviation between the measured values and target values fall within an allowable tolerance (step <b>507</b>). In a preferred embodiment, the target luminance level corresponds to that on a luminance curve having a predetermined target gamma value, e.g., of about 2.2. The target chromaticity point is preferably a white point corresponding to a particular color temperature, e.g., of about 6500K. Different target luminance curves and different target chromaticity points can be selected in order to calibrate the display device <b>10</b> to suit different viewing environments and input video signal characteristics. For example, the target chromaticity point can be selected to correspond to a warmer or cooler color temperature so that the perceived image quality is maintained in warmer or cooler ambient lighting conditions.

0048The allowable tolerance for the deviation between measured and target luminance levels is preferably within ±1%, while the allowable tolerance for the deviation between measured and target chromaticity points is preferably within (±0.002, ±0.002). The tolerance values can be increased or decreased depending on the degree of accuracy required. Nevertheless, depending on the quality of the display device itself, small tolerances might not be attainable. In typical cases, with the tolerances suggested above, after the completion of the calibration process, the measured luminance levels can match the targeted levels to within ±1%, while the measured chromaticity points can fall within (±0.002, ±0.002) of the target chromaticity point and the measured color temperature can fall within ±100K of the target color temperature.

0049If the deviations of the measured luminance levels and chromaticity values are within the allowable tolerances (step <b>507</b>), then the display device <b>10</b> satisfies the calibration standards. The values in the intermediate gamma correction LUTs <b>126</b> are outputted and stored in the display device <b>10</b> (step <b>516</b>), and the calibration process is completed.

0050If the deviations of the measured luminance levels and chromaticity values are not within the allowable tolerances (step <b>507</b>), then the LUT value generator <b>124</b> calculates updated values for the intermediate gamma correction LUTs <b>126</b>. First, the LUT value generator <b>124</b> determines the differential change of measured luminance levels Y and the differential change of measured chromaticity values (x, y) when each primary color component value is increased and decreased a unit differential amount (step <b>508</b>). In particular, for each selected gray level, the test pattern controller <b>130</b> generates R, G and B component output values that include a predetermined differential change from the previous corresponding component output values for each selected gray level. These test patterns are displayed by the display device <b>10</b> and the luminance levels and chromaticity values are measured.

0051In one version, the calibration system <b>100</b> determines the following values for each selected gray level k: <br /><i>Y</i>_R+(<i>k</i>)=<i>Y</i>[min(<i>LUT</i>_R(<i>k</i>)+Δ<i>R, M−</i>1), <i>LUT</i>_G(<i>k</i>), <i>LUT</i>_B(<i>k</i>)], 0≦<i>k≦M−</i>1;<br /><i>x</i>_R+(<i>k</i>)=<i>x</i>[min(<i>LUT</i>_R(<i>k</i>)+Δ<i>R, M−</i>1), <i>LUT</i>_G(<i>k</i>), <i>LUT</i>_B(<i>k</i>)], 0≦<i>k≦M−</i>1;<br /><i>y</i>_R+(<i>k</i>)=<i>y</i>[min(<i>LUT</i>_R(<i>k</i>)+Δ<i>R, M−</i>1), <i>LUT</i>_G(<i>k</i>), <i>LUT</i>_B(<i>k</i>)], 0≦<i>k≦M−</i>1;<br /><i>Y</i>_R−(<i>k</i>)=<i>Y</i>[max(<i>LUT</i>_R(<i>k</i>)−Δ<i>R, </i>0), <i>LUT</i>_G(<i>k</i>), <i>LUT</i>_B(<i>k</i>)], 0≦<i>k≦M−</i>1;<br /><i>x</i>_R−(<i>k</i>)=<i>x</i>[max(<i>LUT</i>_R(<i>k</i>)−Δ<i>R, </i>0), <i>LUT</i>_G(<i>k</i>), <i>LUT</i>_B(<i>k</i>)], 0≦<i>k≦M−</i>1;<br /><i>y</i>_R−(<i>k</i>)=<i>y</i>[max(<i>LUT</i>_R(<i>k</i>)−Δ<i>R, </i>0), <i>LUT</i>_G(<i>k</i>), <i>LUT</i>_B(<i>k</i>)], 0≦<i>k≦M−</i>1;<br /><i>Y</i>_G+(<i>k</i>)=<i>Y[LUT</i>_R(<i>k</i>), min(<i>LUT</i>_G(<i>k</i>)+Δ<i>G, M−</i>1), <i>LUT</i>_B(<i>k</i>)], 0≦<i>k≦M−</i>1;<br /><i>x</i>_G+(<i>k</i>)=<i>x[LUT</i>_R(<i>k</i>), min(<i>LUT</i>_G(<i>k</i>)+Δ<i>G, M−</i>1), <i>LUT</i>_B(<i>k</i>)], 0≦<i>k≦M−</i>1;<br /><i>y</i>_G+(<i>k</i>)=<i>y[LUT</i>_R(<i>k</i>), min(<i>LUT</i>_G(<i>k</i>)+Δ<i>G, M−</i>1), <i>LUT</i>_B(<i>k</i>)], 0≦<i>k≦M−</i>1;<br /><i>Y</i>_G−(<i>k</i>)=<i>Y[LUT</i>_R(<i>k</i>), max(<i>LUT</i>_G(<i>k</i>)−Δ<i>G, </i>0), <i>LUT</i>_B(<i>k</i>)], 0≦<i>k≦M−</i>1;<br /><i>x</i>_G−(<i>k</i>)=<i>x[LUT</i>_R(<i>k</i>), max(<i>LUT</i>_G(<i>k</i>)−Δ<i>G, </i>0), <i>LUT</i>_B(<i>k</i>)], 0≦<i>k≦M−</i>1;<br /><i>y</i>_G−(<i>k</i>)=<i>y[LUT</i>_R(<i>k</i>), max(<i>LUT</i>_G(<i>k</i>)−Δ<i>G, </i>0), <i>LUT</i>_B(<i>k</i>)], 0≦<i>k≦M−</i>1;<br /><i>Y</i>_B+(<i>k</i>)=<i>Y[LUT</i>_R(<i>k</i>), <i>LUT</i>_G(<i>k</i>), min(<i>LUT</i>_B(<i>k</i>)+Δ<i>B, M−</i>1)], 0≦<i>k≦M−</i>1;<br /><i>x</i>_B+(<i>k</i>)=<i>x[LUT</i>_R(<i>k</i>), <i>LUT</i>_G(<i>k</i>), min(<i>LUT</i>_B(<i>k</i>)+Δ<i>B, M−</i>1)], 0≦<i>k≦M−</i>1;<br /><i>y</i>_B+(<i>k</i>)=<i>y[LUT</i>_R(<i>k</i>), <i>LUT</i>_G(<i>k</i>), min(<i>LUT</i>_B(<i>k</i>)+Δ<i>B, M−</i>1)], 0≦<i>k≦M−</i>1;<br /><i>Y</i>_B−(<i>k</i>)=<i>Y[LUT</i>_R(<i>k</i>), <i>LUT</i>_G(<i>k</i>), max(<i>LUT</i>_B(<i>k</i>)−Δ<i>B, </i>0)], 0≦<i>k≦M−</i>1;<br /><i>x</i>_B−(<i>k</i>)=<i>x[LUT</i>_R(<i>k</i>), <i>LUT</i>_G(<i>k</i>), max(<i>LUT</i>_B(<i>k</i>)−Δ<i>B, </i>0)], 0≦<i>k≦M−</i>1;<br /><i>y</i>_B−(<i>k</i>)=<i>y[LUT</i>_R(<i>k</i>), <i>LUT</i>_G(<i>k</i>), max(<i>LUT</i>_B(<i>k</i>)−Δ<i>B, </i>0)], 0≦<i>k≦M−</i>1;<br /> where ΔR, ΔG, and ΔB are the predetermined differential change in the R, G, and B component output values, respectively.

0052The measured values at selected gray levels k are then smoothened and interpolated over all the gray levels i, using well known curve-fitting and interpolation operations to reduce the sensitivity of the calibration algorithm to small errors in the measured data. The smoothened functions are then used to calculate the differential change in luminance level and in chromaticity values due to a unit differential change in each of the primary color components: <br /><i>dY</i>_R(<i>i</i>)=[<i>Y</i>_SR+(<i>i</i>)−<i>Y</i>_SR−(<i>i</i>)]/[min(<i>LUT</i>_R(<i>i</i>)+Δ<i>R, M−</i>1)−max(<i>LUT</i>_R(<i>i</i>)−Δ<i>R, </i>0)],<br /><i>dY</i>_G(<i>i</i>)=[<i>Y</i>_SG+(<i>i</i>)−<i>Y</i>_SG−(<i>i</i>)]/[min(<i>LUT</i>_G(<i>i</i>)+Δ<i>G, M−</i>1)−max(<i>LUT</i>_G(<i>i</i>)−Δ<i>G, </i>0)],<br /><i>dY</i>_B(<i>i</i>)=[<i>Y</i>_SB+(<i>i</i>)−<i>Y</i>_SB−(<i>i</i>)]/[min(<i>LUT</i>_B(<i>i</i>)+Δ<i>B, M−</i>1)−max(<i>LUT</i>_B(<i>i</i>)−Δ<i>B, </i>0)],<br /><i>dx</i>_R(<i>i</i>)=[<i>x</i>_SR+(<i>i</i>)−<i>x</i>_SR−(<i>i</i>)]/[min(<i>LUT</i>_R(<i>i</i>)+Δ<i>R, M−</i>1)−max(<i>LUT</i>_R(<i>i</i>)−Δ<i>R, </i>0)],<br /><i>dx</i>_G(<i>i</i>)=[<i>x</i>_SG+(<i>i</i>)−<i>x</i>_SG−(<i>i</i>)]/[min(<i>LUT</i>_G(<i>i</i>)+Δ<i>G, M−</i>1)−max(<i>LUT</i>_G(<i>i</i>)−Δ<i>G, </i>0)],<br /><i>dx</i>_B(<i>i</i>)=[<i>x</i>_SB+(<i>i</i>)−<i>x</i>_SB−(<i>i</i>)]/[min(<i>LUT</i>_B(<i>i</i>)+Δ<i>B, M−</i>1)−max(<i>LUT</i>_B(<i>i</i>)−Δ<i>B, </i>0)],<br /><i>dy</i>_R(<i>i</i>)=[<i>y</i>_SR+(<i>i</i>)−<i>y</i>_SR−(<i>i</i>)]/[min(<i>LUT</i>_R(<i>i</i>)+Δ<i>R, M−</i>1)−max(<i>LUT</i>_R(<i>i</i>)−Δ<i>R, </i>0)],<br /><i>dy</i>_G(<i>i</i>)=[<i>y</i>_SG+(<i>i</i>)−<i>y</i>_SG−(<i>i</i>)]/[min(<i>LUT</i>_G(<i>i</i>)+Δ<i>G, M−</i>1)−max(<i>LUT</i>_G(<i>i</i>)−Δ<i>G, </i>0)],<br /><i>dy</i>_B(<i>i</i>)=[<i>y</i>_SB+(<i>i</i>)−<i>y</i>_SB−(<i>i</i>)]/[min(<i>LUT</i>_B(<i>i</i>)+Δ<i>B, M−</i>1)−max(<i>LUT</i>_B(<i>i</i>)−Δ<i>B, </i>0)].<br /> where Y_SR+(i), for all the gray levels i, is the smoothened functions corresponding to Y_R+(k), for the selected gray levels k, so are the relationships between the other smoothened and measured quantities.

0053The differential changes in luminance levels and chromaticity values above can be used to define gradient vectors associated with each primary color component. That is: <br /><i>dR</i>(<i>i</i>)=<i>dx</i>_R(<i>i</i>)<i>x+dy</i>_R(<i>i</i>)<i>y, </i><br /><i>dG</i>(<i>i</i>)=<i>dx</i>_G(<i>i</i>)<i>x+dy</i>_G(<i>i</i>)<i>y, </i><br /><i>dB</i>(<i>i</i>)=<i>dx</i>_B(<i>i</i>)<i>x+dy</i>_B(<i>i</i>)<i>y, </i><br /> where x and y are unit vectors on the chromaticity diagram along the x-axis and the y-axis, respectively. Each gradient vector describes how a chromaticity point will move due to a unit differential change in the associated primary color component value.

0054The gradient vectors are then used to determine the differential change needed in one or more primary color component values in order to move the measured chromaticity point on the chromaticity diagram to the target chromaticity point (step <b>510</b>). In one version, the LUT value generator <b>124</b> calculates the minimum-norm vectors that can move the measured chromaticity point (x_S, y_S) on the chromaticity diagram to the target point (x_T, y_T) for every gray level, i, above the threshold gray level. Let dR(i), dG(i), and dB(i) be the differential changes in each of the three primary color component values at gray level i, respectively. The following two equations describe moving the measured chromaticity point [x_S(i), y_S(i)] to the predetermined target point (x_T, y_T): <br /><i>dx</i>_R(<i>i</i>)<i>dR</i>(<i>i</i>)+<i>dx</i>_G(<i>i</i>)<i>dG</i>(<i>i</i>)+<i>dx</i>_B(<i>i</i>)<i>dB</i>(<i>i</i>)=<i>C</i>(<i>i</i>)[<i>x</i>_T<i>−x</i>_S(<i>i</i>)],<br /><i>dy</i>_R(<i>i</i>)<i>dR</i>(<i>i</i>)+<i>dy</i>_G(<i>i</i>)<i>dG</i>(<i>i</i>)+<i>dy</i>_B(<i>i</i>)<i>dB</i>(<i>i</i>)=<i>C</i>(<i>i</i>)[<i>y</i>_T<i>−y</i>_S(<i>i</i>)],<br /> where C(i) is a predetermined chromaticity adjustment curve that smoothly fades out the adjustment at the low gray levels, i.e., C(0)=0 and C(M−1)=1. For such an underdetermined system of linear equations, an infinite number of solutions exist. In a preferred embodiment, the unique solution [dR*(i), dG*(i), dB*(i)] that has a minimum squared norm among all valid solutions is selected.

0055Let P₁(i)=[dR₁(i), dG₁(i), dB₁(i)] and P₂(i)=[dR₂(i), dG₂(i), dB₂(i)] be two solutions where one of the different variables in each solution is zero, e.g., dR₁(i)=0 and dG₂(i)=0. The optimal solution point P*(i)=[dR*(i), dG*(i), dB*(i)] on a solution line including P₁(i) and P₂(i) is that point nearest to the origin. According to the orthogonality principle, a vector from the origin to the optimal solution P*(i) should be orthogonal to a vector along the solution line. Using this principle, the optimal solution point is: <br /><i>P</i>*(<i>i</i>)=[<i>dR</i>*(<i>i</i>), <i>dG</i>*(<i>i</i>), <i>dB</i>*(<i>i</i>)]=<i>P</i>₁(<i>i</i>)+α*(<i>i</i>)[<i>P</i>₂(<i>i</i>)−<i>P</i>₁(<i>i</i>)], i.e.,<br /><i>dR</i>*(<i>i</i>)=<i>dR</i>₁(<i>i</i>)+α*(<i>i</i>)[<i>dR</i>₂(<i>i</i>)−<i>dR</i>₁(<i>i</i>)],<br /><i>dG</i>*(<i>i</i>)=<i>dG</i>₁(<i>i</i>)+α*(<i>i</i>)[<i>dG</i>₂(<i>i</i>)−<i>dG</i>₁(<i>i</i>)],<br /><i>dB</i>*(<i>i</i>)=<i>dB</i>₁(<i>i</i>)+α*(<i>i</i>)[<i>dB</i>₂(<i>i</i>)−<i>dB</i>₁(<i>i</i>)],<br /> where α*(i) is the orthogonality factor:

0056<maths id="MATH-US-00001" num="00001"><math overflow="scroll"><mtable><mtr><mtd><mrow><mrow><msup><mi>α</mi><mo>*</mo></msup><mo>⁡</mo><mrow><mo>(</mo><mi>i</mi><mo>)</mo></mrow></mrow><mo>=</mo><mi /><mo>⁢</mo><mrow><mrow><msub><mi>P</mi><mn>1</mn></msub><mo>⁡</mo><mrow><mo>(</mo><mi>i</mi><mo>)</mo></mrow></mrow><mo>·</mo><mrow><mrow><mo>[</mo><mrow><mrow><msub><mi>P</mi><mn>1</mn></msub><mo>⁡</mo><mrow><mo>(</mo><mi>i</mi><mo>)</mo></mrow></mrow><mo>-</mo><mrow><msub><mi>P</mi><mn>2</mn></msub><mo>⁡</mo><mrow><mo>(</mo><mi>i</mi><mo>)</mo></mrow></mrow></mrow><mo>]</mo></mrow><mo>/</mo><msup><mrow><mo></mo><mrow><mrow><msub><mi>P</mi><mn>1</mn></msub><mo>⁡</mo><mrow><mo>(</mo><mi>i</mi><mo>)</mo></mrow></mrow><mo>-</mo><mrow><msub><mi>P</mi><mn>2</mn></msub><mo>⁡</mo><mrow><mo>(</mo><mi>i</mi><mo>)</mo></mrow></mrow></mrow><mo></mo></mrow><mn>2</mn></msup></mrow></mrow></mrow></mtd></mtr><mtr><mtd><mrow><mo>=</mo><mi /><mo>⁢</mo><mrow><mo>{</mo><mrow><mrow><mrow><msub><mi>dR</mi><mn>1</mn></msub><mo>⁡</mo><mrow><mo>(</mo><mi>i</mi><mo>)</mo></mrow></mrow><mo>⁡</mo><mrow><mo>[</mo><mrow><mrow><msub><mi>dR</mi><mn>1</mn></msub><mo>⁡</mo><mrow><mo>(</mo><mi>i</mi><mo>)</mo></mrow></mrow><mo>-</mo><mrow><msub><mi>dR</mi><mn>2</mn></msub><mo>⁡</mo><mrow><mo>(</mo><mi>i</mi><mo>)</mo></mrow></mrow></mrow><mo>]</mo></mrow></mrow><mo>+</mo><mrow><mrow><msub><mi>dG</mi><mn>1</mn></msub><mo>⁡</mo><mrow><mo>(</mo><mi>i</mi><mo>)</mo></mrow></mrow><mo>⁡</mo><mrow><mo>[</mo><mrow><mrow><msub><mi>dG</mi><mn>1</mn></msub><mo>⁡</mo><mrow><mo>(</mo><mi>i</mi><mo>)</mo></mrow></mrow><mo>-</mo><mrow><msub><mi>dG</mi><mn>2</mn></msub><mo>⁡</mo><mrow><mo>(</mo><mi>i</mi><mo>)</mo></mrow></mrow></mrow><mo>]</mo></mrow></mrow><mo>+</mo></mrow></mrow></mrow></mtd></mtr><mtr><mtd><mrow><mrow><mi /><mo>⁢</mo><mrow><mrow><mo>+</mo><mi>d</mi></mrow><mo>⁢</mo><mstyle><mspace width="0.3em" height="0.3ex" /></mstyle><mo>⁢</mo><mrow><mrow><msub><mi>B</mi><mn>1</mn></msub><mo>⁡</mo><mrow><mo>(</mo><mi>i</mi><mo>)</mo></mrow></mrow><mo>⁡</mo><mrow><mo>[</mo><mrow><mrow><mi>d</mi><mo>⁢</mo><mstyle><mspace width="0.3em" height="0.3ex" /></mstyle><mo>⁢</mo><mrow><msub><mi>B</mi><mn>1</mn></msub><mo>⁡</mo><mrow><mo>(</mo><mi>i</mi><mo>)</mo></mrow></mrow></mrow><mo>-</mo><mrow><mi>d</mi><mo>⁢</mo><mstyle><mspace width="0.3em" height="0.3ex" /></mstyle><mo>⁢</mo><mrow><msub><mi>B</mi><mn>2</mn></msub><mo>⁡</mo><mrow><mo>(</mo><mi>i</mi><mo>)</mo></mrow></mrow></mrow></mrow><mo>]</mo></mrow></mrow></mrow><mo>}</mo></mrow><mo>/</mo><mrow><mo>{</mo><mrow><msup><mrow><mo>[</mo><mrow><mrow><msub><mi>dR</mi><mn>1</mn></msub><mo>⁡</mo><mrow><mo>(</mo><mi>i</mi><mo>)</mo></mrow></mrow><mo>-</mo><mrow><msub><mi>dR</mi><mn>2</mn></msub><mo>⁡</mo><mrow><mo>(</mo><mi>i</mi><mo>)</mo></mrow></mrow></mrow><mo>]</mo></mrow><mn>2</mn></msup><mo>+</mo></mrow></mrow></mrow></mtd></mtr><mtr><mtd><mrow><mrow><mi /><mo>⁢</mo><mrow><msup><mrow><mo>[</mo><mrow><mrow><msub><mi>dG</mi><mn>1</mn></msub><mo>⁡</mo><mrow><mo>(</mo><mi>i</mi><mo>)</mo></mrow></mrow><mo>-</mo><mrow><msub><mi>dG</mi><mn>2</mn></msub><mo>⁡</mo><mrow><mo>(</mo><mi>i</mi><mo>)</mo></mrow></mrow></mrow><mo>]</mo></mrow><mn>2</mn></msup><mo>+</mo><msup><mrow><mo>[</mo><mrow><mrow><mi>d</mi><mo>⁢</mo><mstyle><mspace width="0.3em" height="0.3ex" /></mstyle><mo>⁢</mo><mrow><msub><mi>B</mi><mn>1</mn></msub><mo>⁡</mo><mrow><mo>(</mo><mi>i</mi><mo>)</mo></mrow></mrow></mrow><mo>-</mo><mrow><mi>d</mi><mo>⁢</mo><mstyle><mspace width="0.3em" height="0.3ex" /></mstyle><mo>⁢</mo><mrow><msub><mi>B</mi><mn>2</mn></msub><mo>⁡</mo><mrow><mo>(</mo><mi>i</mi><mo>)</mo></mrow></mrow></mrow></mrow><mo>]</mo></mrow><mn>2</mn></msup></mrow><mo>}</mo></mrow><mo>.</mo></mrow></mtd></mtr></mtable></math></maths><img file="US7495679B2_D0001.tif" />

0057Once the differential changes to one or more of the primary color component values are determined for each gray level, updated values for each of the intermediate gamma correction LUTs <b>126</b> are calculated for each gray level (step <b>512</b>) according to the following expressions:

0000Note that LUT_R(i), LUT_G(i), and LUT_B(i) at the right of the equal sign are the old values, while those at the left are the corresponding updated values. <br /><i>LUT</i>_R(<i>i</i>)=max{min[<i>LUT</i>_R(<i>i</i>)+<i>dR</i>*(<i>i</i>), <i>M−</i>1], 0},<br /><i>LUT</i>_G(<i>i</i>)=max{min[<i>LUT</i>_G(<i>i</i>)+<i>dG</i>*(<i>i</i>), <i>M−</i>1], 0},<br /><i>LUT</i>_B(<i>i</i>)=max{min[<i>LUT</i>_B(<i>i</i>)+<i>dB</i>*(<i>i</i>), <i>M−</i>1], 0},<br /> In a preferred embodiment, updated luminance levels are also calculated according to the following expression, where Y_S(i) at the right of the equal sign is the old value, and that at the left is the updated value: <br /><i>Y</i>_S(<i>i</i>)=<i>Y</i>_S(<i>i</i>)+<i>dY</i>_R(<i>i</i>)<i>dR</i>*(<i>i</i>)+<i>dY</i>_G(<i>i</i>)<i>dG</i>*(<i>i</i>)+<i>dY</i>_B(<i>i</i>)<i>dB</i>*(<i>i</i>).<br /><figref idref="DRAWINGS">FIG. 7A</figref> is a graph showing the updated values of the intermediate gamma correction LUTs <b>126</b>, i.e., LUT_R(i), LUT_G(i), LUT_B(i), versus gray level and <figref idref="DRAWINGS">FIG. 7B</figref> is a graph showing the updated luminance levels Y_S(i) versus gray level. In <figref idref="DRAWINGS">FIG. 7A</figref>, the updated values LUT_R(i), LUT_G(i), LUT_B(i) noticeably deviate from the initial linear function above the threshold gray level of about 10% to 20% of the maximum gray level, which is indicative of the degree to which the input video signal should be adjusted in order to substantially eliminate gray-level color deviations of the display device <b>10</b>.

0058In <figref idref="DRAWINGS">FIG. 7A</figref>, at higher gray level values, the LUT values for at least one of the primary color components begin to saturate (at a maximum value of M−1) at a gray level N less than the maximum gray level, M−1. Similarly, in <figref idref="DRAWINGS">FIG. 7B</figref>, at higher gray level values, the luminance levels Y_S(i) increase smoothly and reach a maximum luminance value Y_maxat a gray level N, which is less than the maximum gray level, M−1. Smooth increasing of the luminance levels Y_S(i) beyond the maximum luminance value Y_maxis not attainable by the display screen <b>14</b> because at least one of the primary color components saturate beyond the gray level N. In order to match desired chromaticity values and luminance levels for each of the gray levels, including those beyond the gray level N, the values in the intermediate gamma correction LUTs <b>126</b> are adjusted (step <b>514</b>).

0059In one version, the LUT values are adjusted in the following manner. Firstly, a maximum luminance level Y_MAXis calculated at a gray level N where the LUT values for at least one of the intermediate gamma correction LUTs <b>126</b> start to saturate at the maximum value of M−1. In addition, a minimum luminance level Y_MINis calculated where the LUT values for all the intermediate gamma correction LUTs <b>126</b> equal zero (0). Secondly, for each valid index i of the intermediate gamma correction LUTs <b>126</b>, set a target luminance value Y_T(i) to: <br /><i>Y</i>_T(<i>i</i>)=<i>L</i>(<i>i</i>)[(<i>Y</i>_MAX<i>−Y</i>_MIN)γ_T(<i>i</i>)+<i>Y</i>_MIN]+[1−<i>L</i>(<i>i</i>)]<i>Y</i>_S(<i>i</i>),<br /> where L(i) is a predetermined luminance adjustment curve that smoothly fades out the adjustment at the low gray level portion, i.e., L(0)=0 and L(M−1)=1, and γ_T(i) is a normalized target luminance curve typically represented by a gamma function with a predetermined gamma value, i.e., γ_T(i)=[i/(M−1)]^γ, i=0, 1, . . . , M−1. Thirdly, for each valid index i of the intermediate gamma correction LUTs <b>126</b>, search an index j such that: <br /><i>Y</i>_S(<i>j</i>)≦<i>Y</i>_T(<i>j</i>)&lt;<i>Y</i>_S(<i>j+</i>1)<br /> and calculate a fractional adjustment factor: <br /><i>F</i>(<i>i</i>)=[<i>Y</i>_T(<i>i</i>)−<i>Y</i>_S(<i>j</i>)]/[<i>Y</i>_S(<i>j+</i>1)−<i>Y</i>_S(<i>j</i>)].<br /> The values in each of the intermediate gamma correction LUTs <b>126</b> and the luminance levels are then adjusted by the following expressions, where LUT_R(i), LUT_G(i), LUT_B(i), and Y_S(i) at the right of the equal sign are the updated values, while those at the left are the corresponding adjusted values: <br /><i>LUT</i>_R(<i>i</i>)=round[(1−<i>F</i>(<i>i</i>))<i>LUT</i>_R(<i>j</i>)+<i>F</i>(<i>i</i>)<i>LUT</i>_R(<i>j+</i>1)],<br /><i>LUT</i>_G(<i>i</i>)=round[(1−<i>F</i>(<i>i</i>))<i>LUT</i>_G(<i>j</i>)+<i>F</i>(<i>i</i>)<i>LUT</i>_G(<i>j+</i>1)],<br /><i>LUT</i>_B(<i>i</i>)=round[(1−<i>F</i>(<i>i</i>))<i>LUT</i>_B(<i>j</i>)+<i>F</i>(<i>i</i>)<i>LUT</i>_B(<i>j+</i>1)],<br /><i>Y</i>_S(<i>i</i>)=[1−<i>F</i>(<i>i</i>)]<i>Y</i>_S(<i>j</i>)+<i>F</i>(<i>i</i>)<i>Y</i>_S(<i>j+</i>1),<br /> where round[•] denotes rounding to the nearest integer number. <figref idref="DRAWINGS">FIG. 8A</figref> and <figref idref="DRAWINGS">FIG. 8B</figref> are graphs showing the adjusted values of the intermediate gamma correction LUTs <b>126</b> and luminance levels, respectively. At this stage in the calibration process, both the target chromaticity point and the target luminance curve are matched simultaneously.

0060After the intermediate gamma correction LUT values and luminance levels have been simultaneously adjusted (step <b>514</b>), the calibration module <b>120</b> determines whether calibration process is done (step <b>515</b>). For example, the number of times the calibration system <b>100</b> performs steps <b>502</b> through <b>514</b> can be set not to exceed a specified number. If the calibration process is not done (step <b>515</b>), e.g., the number of iterations does not exceed the maximum number, the calibration system <b>100</b> repeats steps <b>502</b> through <b>514</b>. In this iteration, the input test patterns to the display screen <b>14</b> are adjusted by the values in each of the intermediate gamma correction LUTs <b>126</b> and the displayed test patterns by the display screen <b>14</b> correspond to the adjusted primary color components.

0061If the calibration process is done (step <b>515</b>), e.g., because the number of iterations equals the maximum number of iterations, the values in each of the intermediate gamma correction LUTs <b>126</b> are outputted and stored in the storage mechanism 16 of the display device <b>10</b> (step <b>516</b>), and the calibration process is completed. In one version, before outputting the LUT values, the calibration module <b>120</b> can check slope, continuity, and monotonic properties for the LUT values of the intermediate gamma correction LUTs <b>126</b> throughout all valid indexes, and make necessary modifications if needed. In operation, the display device <b>10</b> can load the LUT values from the storage mechanism <b>16</b> into the embedded gamma correction LUTs <b>12</b>.

0062<figref idref="DRAWINGS">FIG. 9</figref> is a graph showing the chromaticity trajectory after the calibration process. Comparing <figref idref="DRAWINGS">FIG. 9</figref> to <figref idref="DRAWINGS">FIG. 4</figref>, after calibration, the measured chromaticity points fall on or near the target chromaticity point. Accordingly, the gray-level color temperature shift typically exhibited by a typical uncalibrated display device <b>10</b> (<figref idref="DRAWINGS">FIG. 4</figref>) is substantially eliminated after the calibration process according to the present invention.

0063In the calibration process described above, the chromaticity values are fixed to a target chromaticity point (x_T, y_T) and the gamma value of the luminance curve is fixed to a target gamma value γ_Tthroughout all gray levels above the threshold gray level. In another version, the chromaticity values can be fixed to a target chromaticity function [x_T(i), y_T(i)] and the luminance levels can be fixed to a target luminance level function Y_T(i).

0064<figref idref="DRAWINGS">FIG. 10</figref> is a flowchart illustrating a calibration process according to this version of the present invention. Process steps <b>900</b> to <b>908</b> described in <figref idref="DRAWINGS">FIG. 10</figref> are identical to process steps <b>500</b> through <b>508</b> in <figref idref="DRAWINGS">FIG. 5</figref>, and therefore, the discussion above relating to steps <b>500</b> through <b>508</b> will not be repeated here.

0065After step <b>908</b>, the differential changes of luminance levels and chromaticity values with respect to a unit differential change in each of the primary color components are used to define gradient vectors representing how a measured point [Y_S(i), x_S(i), y_S(i)] on a luminance-chromaticity diagram will move due to a unit differential change in one of the primary color component values. <figref idref="DRAWINGS">FIG. 11</figref> illustrates a three-dimensional luminance-chromaticity diagram showing the measured point [Y_S(i), x_S(i), y_S(i)] and a predetermined target point [Y_T(j), x_T(j), y_T(j)]. The gradient vectors are: <br /><i>dR</i>(<i>i</i>)=<i>dY</i>_R(<i>i</i>)<i>Y+dx</i>_R(<i>i</i>)<i>x+dy</i>_R(<i>i</i>)<i>y, </i><br /><i>dG</i>(<i>i</i>)=<i>dY</i>_G(<i>i</i>)<i>Y+dx</i>_G(<i>i</i>)<i>x+dy</i>_G(<i>i</i>)<i>y, </i><br /><i>dB</i>(<i>i</i>)=<i>dY</i>_B(<i>i</i>)<i>Y+dx</i>_B(<i>i</i>)<i>x+dy</i>_B(<i>i</i>)<i>y, </i><br /> where Y is the unit vector along the Y-axis, and x and y are unit vectors on the chromaticity diagram along the x-axis and the y-axis, respectively.

0066The gradient vectors are then used to determine the differential change needed in one or more primary color component values in order to move the measured luminance-chromaticity point to the target luminance-chromaticity point (step <b>910</b>). In this version, the LUT value generator <b>124</b> calculates the minimum-norm vectors that can move the measured luminance-chromaticity point [Y_S(i), x_S(i), y_S(i)] on the luminance-chromaticity diagram to the target point [Y_T(j), x_T(j), y_T(j)] for every gray level, i, above the threshold gray level. If dR(i, j), dG(i, j), and dB(i, j) are the differential changes in the values of each of the intermediate gamma correction LUTs <b>126</b> at index i, respectively, in order to move the measured luminance-chromaticity point [Y_S(i), x_S(i), y_S(i)] to the target point [Y_T(j), x_T(j), y_T(j)], then the following three equations apply: <br /><i>dY</i>_R(<i>i</i>)<i>dR</i>(<i>i, j</i>)+<i>dY</i>_G(<i>i</i>)<i>dG</i>(<i>i, j</i>)+<i>dY</i>_B(<i>i</i>)<i>dB</i>(<i>i, j</i>)=<i>L</i>(<i>j</i>)<i>Y</i>_T(<i>j</i>)+[1−<i>L</i>(<i>j</i>)]<i>Y</i>_S(<i>j</i>)−<i>Y</i>_S(<i>i</i>),<br /><i>dx</i>_R(<i>i</i>)<i>dR</i>(<i>i, j</i>)+<i>dx</i>_G(<i>i</i>)<i>dG</i>(<i>i, j</i>)+<i>dx</i>_B(<i>i</i>)<i>dB</i>(<i>i, j</i>)=<i>C</i>(<i>j</i>)<i>x</i>_T(<i>j</i>)+[1−<i>C</i>(<i>j</i>)]<i>x</i>_S(<i>j</i>)−<i>x</i>_S(<i>i</i>),<br /><i>dy</i>_R(<i>i</i>)<i>dR</i>(<i>i, j</i>)+<i>dy</i>_G(<i>i</i>)<i>dG</i>(<i>i, j</i>)+<i>dy</i>_B(<i>i</i>)<i>dB</i>(<i>i, j</i>)=<i>C</i>(<i>j</i>)<i>y</i>_T(<i>j</i>)+[1−<i>C</i>(<i>j</i>)]<i>y</i>_S(<i>j</i>)−<i>y</i>_S(<i>i</i>),<br /> where L(i) is a predetermined luminance adjustment curve that smoothly fades out the adjustment at the low gray levels, i.e., L(0)=0 and L(M−1)=1, and C(i) is a predetermined chromaticity adjustment curve that smoothly fades out the adjustment at the low gray levels, i.e., C(0)=0 and C(M−1)=1.

0067For each given i and j index, a unique solution exists. Thus, for each measured point [Y_S(i), x_S(i), y_S(i)], the unique solution [dR*(i), dG*(i), dB*(i)] is calculated that has a minimum squared norm among all valid index j of a predetermined target point [Y_T(j), x_T(j), y_T(j)], namely: <br />[<i>dR</i>*(<i>i</i>), <i>dG</i>*(<i>i</i>), <i>dB</i>*(<i>i</i>)]=[<i>dR</i>(<i>i, j</i>*), <i>dG</i>(<i>i, j</i>*), <i>dB</i>(<i>i, j</i>*)],<br /> where j* is the optimal index of the optimal predetermined target point [Y_T(j*), x_T(j*), y_T(j*)] such that: <br />[<i>dR</i>(<i>i, j</i>*)]²<i>+[dG</i>(<i>i, j</i>*)]²<i>+[dB</i>(<i>i, j</i>*)]²<i>≦[dR</i>(<i>i, j</i>)]²<i>+[dG</i>(<i>i, j</i>)]²<i>+[dB</i>(<i>i, j</i>)]²,<br /> for all valid index j of a predetermined target point except j*. If more than one j* exists that achieves the minimum squared norm, the j* that is nearest to i is selected.

0068Once the differential changes to one or more of the primary color component values is determined for each gray level, updated values for each of the intermediate gamma correction LUTs <b>126</b>, for the luminance levels, and for chromaticity values are calculated for each gray level (step <b>912</b>) according to the following expressions, where LUT_R(i), LUT_G(i), LUT_B(i), Y_S(i), x_S(i), and y_S(i) at the right of the equal sign are the old values, and those at the left are the corresponding updated values: <br /><i>LUT</i>_R(<i>i</i>)=max{min[<i>LUT</i>_R(<i>i</i>)+<i>dR</i>*(<i>i</i>), <i>M−</i>1], 0},<br /><i>LUT</i>_G(<i>i</i>)=max{min[<i>LUT</i>_G(<i>i</i>)+<i>dG</i>*(<i>i</i>), <i>M−</i>1], 0},<br /><i>LUT</i>_B(<i>i</i>)=max{min[<i>LUT</i>_B(<i>i</i>)+<i>dB</i>*(<i>i</i>), <i>M−</i>1], 0},<br /><i>Y</i>_S(<i>i</i>)=<i>Y</i>_S(<i>i</i>)+<i>dY</i>_R(<i>i</i>)<i>dR</i>*(<i>i</i>)+<i>dY</i>_G(<i>i</i>)<i>dG</i>*(<i>i</i>)+<i>dY</i>_B(<i>i</i>)<i>dB</i>*(<i>i</i>),<br /><i>x</i>_S(<i>i</i>)=<i>x</i>_S(<i>i</i>)+<i>dx</i>_R(<i>i</i>)<i>dR</i>*(<i>i</i>)+<i>dx</i>_G(<i>i</i>)<i>dG</i>*(<i>i</i>)+<i>dx</i>_B(<i>i</i>)<i>dB</i>*(<i>i</i>),<br /><i>y</i>_S(<i>i</i>)=<i>y</i>_S(<i>i</i>)+<i>dy</i>_R(<i>i</i>)<i>dR</i>*(<i>i</i>)+<i>dy</i>_G(<i>i</i>)<i>dG</i>*(<i>i</i>)+<i>dy</i>_B(<i>i</i>)<i>dB</i>*(<i>i</i>).<br /> In addition, the optimal index j* of the optimal predetermined target point for all valid indexes i of those tables is recorded as:

0069<maths id="MATH-US-00002" num="00002"><math overflow="scroll"><mrow><mrow><msub><mi>J</mi><mi>OPT</mi></msub><mo>⁡</mo><mrow><mo>(</mo><mi>i</mi><mo>)</mo></mrow></mrow><mo>=</mo><mrow><mi>arg</mi><mo>⁢</mo><mstyle><mspace width="0.8em" height="0.8ex" /></mstyle><mo>⁢</mo><mrow><munder><mi>min</mi><mi>j</mi></munder><mo>⁢</mo><mrow><mo>{</mo><mrow><msup><mrow><mo>[</mo><mrow><mi>dR</mi><mo>⁡</mo><mrow><mo>(</mo><mrow><mi>i</mi><mo>,</mo><mi>j</mi></mrow><mo>)</mo></mrow></mrow><mo>]</mo></mrow><mn>2</mn></msup><mo>+</mo><msup><mrow><mo>[</mo><mrow><mi>dG</mi><mo>⁡</mo><mrow><mo>(</mo><mrow><mi>i</mi><mo>,</mo><mi>j</mi></mrow><mo>)</mo></mrow></mrow><mo>]</mo></mrow><mn>2</mn></msup><mo>+</mo><msup><mrow><mo>[</mo><mrow><mi>dB</mi><mo>⁡</mo><mrow><mo>(</mo><mrow><mi>i</mi><mo>,</mo><mi>j</mi></mrow><mo>)</mo></mrow></mrow><mo>]</mo></mrow><mn>2</mn></msup></mrow><mo>}</mo></mrow></mrow></mrow></mrow></math></maths><img file="US7495679B2_D0002.tif" />

0070In order to match desired chromaticity functions [x_T(j), y_T(j)] and the luminance function Y_T(j) for each of the gray levels, the values in the intermediate gamma correction LUTs <b>126</b> are adjusted (step <b>914</b>). In this version, the LUT values are adjusted in the following manner.

0071Firstly, the maximum index j_MAXof the predetermined target point that can be matched when the updated values for at least one of the intermediate gamma correction tables <b>126</b> start to saturate at a maximum value of M−1 is determined. Secondly, for each index i of the intermediate gamma correction tables <b>126</b>, an adjusted index j_Tof the predetermined target point is set in the following manner: <br /><i>J</i>_T(<i>i</i>)=<i>i·j</i>_MAX/(<i>M−</i>1).<br /> Thirdly, for each index i of the intermediate gamma correction tables 126, indices k₁and k₂are identified such that J_OPT(k₁) and J_OPT(k₂) have minimum difference among all pairs of indices k₁and k₂that can satisfy the following condition: <br /><i>J</i>_OPT(<i>k</i>₁)≦<i>J</i>_T(<i>i</i>)≦<i>J</i>_OPT(<i>k</i>₂).<br /> The fractional adjustment factor is calculated: <br /><i>F</i>(<i>i</i>)=[<i>J</i>_T(<i>i</i>)−<i>J</i>_OPT(<i>k</i>₁)]/[<i>J</i>_OPT(<i>k</i>₂)−<i>J</i>_OPT(<i>k</i>₁)],<br /> and then values for the intermediate gamma correction LUTs <b>126</b>, the luminance levels, and the chromaticity values throughout all valid indexes of those tables are then adjusted according to the following expressions, where LUT_R(i), LUT_G(i), LUT_B(i), Y_S(i), x_S(i), and y_S(i) at the right of the equal sign are the updated values, and those at the left are the corresponding adjusted values. <br /><i>LUT</i>_R(<i>i</i>)=round[(1−<i>F</i>(<i>i</i>))<i>LUT</i>_R(<i>k</i>₁)+<i>F</i>(<i>i</i>)<i>LUT</i>_R(<i>k</i>₂)],<br /><i>LUT</i>_G(<i>i</i>)=round[(1−<i>F</i>(<i>i</i>))<i>LUT</i>_G(<i>k</i>₁)+<i>F</i>(<i>i</i>)<i>LUT</i>_G(<i>k</i>₂)],<br /><i>LUT</i>_B(<i>i</i>)=round[(1−<i>F</i>(<i>i</i>))<i>LUT</i>_B(<i>k</i>₁)+<i>F</i>(<i>i</i>)<i>LUT</i>_B(<i>k</i>₂)],<br /><i>Y</i>_S(<i>i</i>)=[1−<i>F</i>(<i>i</i>)]<i>Y</i>_S(<i>k</i>₁)+<i>F</i>(<i>i</i>)<i>Y</i>_S(<i>k</i>₂),<br /><i>x</i>_S(<i>i</i>)=[1−<i>F</i>(<i>i</i>)]<i>x</i>_S(<i>k</i>₁)+<i>F</i>(<i>i</i>)<i>x</i>_S(<i>k</i>₂),<br /><i>y</i>_S(<i>i</i>)=[1−<i>F</i>(<i>i</i>)]<i>y</i>_S(<i>k</i>₁)+<i>F</i>(<i>i</i>)<i>y</i>_S(<i>k</i>₂).

0072After the LUT values, luminance levels and chromaticity values have been simultaneously adjusted (step <b>914</b>), the calibration module <b>120</b> determines whether calibration process is done (step <b>915</b>), as described above. If the calibration process is not done, the calibration system <b>100</b> repeats steps <b>902</b> through <b>914</b>. If the calibration process is done (step <b>915</b>), the values in each of the intermediate gamma correction LUTs <b>126</b> are outputted and stored in the storage mechanism <b>16</b> of the display device <b>10</b> (step <b>916</b>), and the calibration process is completed. In operation, the display device <b>10</b> can load the LUT values from the storage mechanism <b>16</b> into the embedded gamma correction LUTs <b>12</b>.

0073As briefly mentioned above, the calibration system <b>100</b> according to the present invention can generate a plurality of sets of LUT values that calibrate the display device <b>10</b> for different viewing conditions and input video signal characteristics. Each set of LUT values can correspond to a different gamma value, different color chromaticity values, and/or a different color temperature. Each set can be stored in the storage mechanism <b>16</b>. Thus, depending on the preferences of the viewer or on the viewing environment and input video signal characteristics, an appropriate set of LUT values can be selected and loaded into the embedded gamma correction LUTs <b>12</b> to achieve optimal perceived image quality.

0074Referring again to <figref idref="DRAWINGS">FIG. 3</figref>, the embedded gamma correction LUTs <b>12</b> reside in the display device <b>10</b>, e.g., LCD, CRT, or PDP. The display device <b>10</b> is utilized by a display system (not shown), such as a television, which receives and processes an input video signal, and passes the input video signal to the display device <b>10</b> where the embedded gamma correction LUTs <b>12</b> are used to adjust the input video signal before it is displayed on the display screen <b>14</b>.

0075In one version, the embedded gamma correction LUTs <b>12</b> can reside in the display system. <figref idref="DRAWINGS">FIG. 12</figref> depicts an exemplary display system according to a version of the present invention. The display system <b>20</b> includes a signal receiving unit <b>22</b> that is coupled to a tuner box <b>24</b>, and a video decoder <b>28</b>. Incoming signals <b>21</b>, such as television signals, are captured by the signal receiving unit <b>22</b> and transmitted to the tuner box <b>24</b>. The tuner box <b>24</b> includes a converter <b>25</b> and a demodulation unit <b>26</b> that transforms the incoming signal <b>21</b> into an analog signal <b>27</b>. The analog signal <b>27</b> is received by the video decoder <b>28</b>, which outputs an interlaced video signal <b>29</b>. A video processor module <b>30</b> converts the interlaced video signal <b>29</b> into a progressive video signal <b>32</b>. The progressive video signal <b>32</b> is then inputted into the embedded gamma correction LUTs <b>12</b> for adjustment and correction. The adjusted progressive video signal <b>36</b> is then displayed via the display device <b>34</b>, such as an LCD, CRT or PDP.

0076In another version, the display screen <b>14</b> can contain multiple display regions with different input video signal characteristics. <figref idref="DRAWINGS">FIG. 13</figref> depicts a display screen <b>14</b><i>a </i>containing two display regions (<b>15</b><i>a, </i><b>15</b><i>b</i>) with different input video signal characteristics. For example, display region A (<b>15</b><i>a</i>) may display an input video signal from a television signal source while display region B (<b>15</b><i>b</i>) may display an input video signal from a computer graphic card.

0077<figref idref="DRAWINGS">FIG. 14</figref> depicts an exemplary display system according to another version of the present invention, where similar components are identified by similar item numerals. The display system <b>20</b><i>a </i>includes a video processor module <b>30</b><i>a </i>that converts a first interlaced video signal <b>29</b> into a progressive video signal. In addition to the first interlaced video signal <b>29</b>, the video processor module <b>30</b><i>a </i>also receives a second input video signal <b>23</b> and converts it into a second progressive video signal. In a preferred embodiment, the video processor module <b>30</b><i>a </i>is configured to merge the two progressive video signals <b>29</b>, <b>23</b> according to the positions of their assigned display regions, e.g., <b>15</b><i>a </i>and <b>15</b><i>b</i>, on the display screen <b>14</b><i>a</i>. The video processor module <b>30</b><i>a </i>then generates a merged progressive video signal <b>32</b><i>a</i>. The merged progressive video signal <b>32</b><i>a </i>is inputted into a plurality of sets of embedded gamma correction LUTs <b>12</b><i>a </i>for adjustment and correction according to the input video signal characteristics.

0078According to this version of the present invention, a display region selection signal <b>33</b> is also generated by the video processor module <b>30</b><i>a </i>and inputted into the plurality of sets of embedded gamma correction LUTs <b>12</b><i>a</i>. The selection signal <b>33</b> determines which one of the sets of embedded gamma correction LUTs <b>12</b><i>a </i>can be used according to the positions of the multiple display regions, e.g., <b>15</b><i>a </i>and <b>15</b><i>b</i>. The adjusted progressive video signal <b>36</b> is then displayed via the display device <b>34</b>, such as an LCD, CRT or PDP.

0079Versions of the present invention provide a method and system that automatically calibrate the luminance and color of a color display. The calibration system and process ensure accurate, stable gamma values, and a smooth luminance curve for typical display devices. In addition, versions of the present invention allow the luminance curve and gamma value to be changed to a setting different from a factory-preset setting, in order to achieve optimum perceived image quality under various viewing conditions and input video signal characteristics. Additional benefits include: (1) increasing the accuracy of color reproduction by reducing gray-level color temperature shift from as much as 5000K for a typical uncalibrated display device to less than 200K variation of color temperature for all gray levels above the threshold; (2) providing precise color balance among red, green, and blue primary color components throughout most of the gray levels; (3) maintaining gray-level color consistency and uniformity; (4) equalizing a color image displayed by a display device based on the color temperature of the ambient light for improved perceived picture quality; (5) adjusting the gamma value of the luminance curve of a display device according to the predetermined gamma value of an input gamma pre-corrected video signal; and (6) allowing the viewer to adjust the gamma value of the luminance curve of a display device based on the brightness level of the viewing environment for optimum image reproduction.

0080The calibration system and process according to the present invention can be used by television manufacturers to calibrate their products easily and quickly to satisfy different requirements relating to the accuracy and specification of luminance and color reproduction. Although different types of display devices, e.g., LCD, CRT, or PDP, can exhibit slightly different luminance and color characteristics, the calibration system according to the present invention can be used to ensure that an end product, e.g., television or monitor, exhibits consistent luminance and color characteristics. Indeed, the calibration system can be part of the factory assembly line. Furthermore, the calibration system according to the present invention provides television manufacturers with the flexibility to readily and conveniently procure key display devices and components from different vendors subject to availability and market conditions without jeopardizing the luminance and color consistency and uniformity of the end products.

0081The present invention has been described with reference to certain preferred versions. Nevertheless, other versions are possible. For example, the number and types of lookup tables can vary. Further, alternative steps equivalent to those described for the luminance and color calibration process can also be used in accordance with the parameters of the described implementations, as would be apparent to one of ordinary skill. Therefore, the spirit and scope of the appended claims should not be limited to the description of the preferred versions contained herein.